The effect of framed labor market messages on training behavior: exploring the role of emotions, beliefs and expectations as well as economic preferences

The information experiment will be conducted online, utilizing a randomized experimental design with one control group and two treatment groups, as detailed below. Participants will be randomly assigned to one of the groups and receive a tailored information intervention as part of the survey:

Control group: no information

Treatment group 1: General labour market information + more specific labour market information highlighting the benefits of engaging in training (gain-framed message)

Treatment group 2: General labour market information + more specific labour market information highlighting the costs of not engaging in training (loss-framed message)

Specifically, in both treatment groups, we provide information on the changes in the labour market. Following this, each treatment group receives framed information on the potential outcomes of training (see Appendix of additional document for detailed framing information).

Gain-framed message (Treatment Group 1):
Highlights the benefits of engaging in training, such as the opportunity to deepen or update knowledge, the chance to work in a different occupation, the prospects for a future wage increase, the potential for increased job satisfaction, and the opportunity to secure a new job.

Loss-framed message (Treatment Group 2):
Emphasizes the costs of not engaging in training, such as missing the chance to expand knowledge, missing opportunities to work in a different occupation, losing the potential for a future salary increase, missing out on the possibility of greater job satisfaction, and the risk of losing the chance for a new job.

2024-12-23

2025-01-31

• The direct effects of the framed labour market information on emotions (e.g., hope, fear), as well as training attitudes and behaviour
• The (mediating) role of labor market beliefs and expectations

Emotional Responses
Literature on message framing suggests that emotional reactions, such as fear or hope, can serve as critical pathways through which interventions influence behaviour (Keller et al., 2003). The survey captures both positive and negative emotional responses:
• Positive emotion: three items measuring feelings of hope (Lu, 2016; Nabi & Myrick, 2019; Tao et al., 2023)
• Negative emotion: three items measuring feelings of fear (Tao et al., 2023; Zheng et al., 2022)

Building on research that emphasizes the importance of cognitive mediators in behavior change, the survey measures beliefs and expectations related to training and the labor market:
• Beliefs about training (perceived need for training (need to update skills), benefits of training, cost-benefit ratio of training, concerns about technological change)
• Labor market expectations (including self-perceived employability, the risk of job loss, the likelihood of voluntarily leaving a position, exiting the workforce, pursuing a career change to an entirely different field or profession, or opting for early retirement)

These constructs are included to explore whether the intervention changes perceptions and whether these shifts mediate behavioral intentions.

Outcome measures focus on the core behavioral and intention-based variables of interest:
• Training information acquisition:
- Participants indicate how likely they are to engage information acquisition about training within the 12 months
- Participants indicate how likely they are to take active steps to explore available training options
- Participants indicate whether they would like to be redirected to a website that gives more information about training
• Training intentions: Participants indicate how likely they are to engage in training activities within the next year.
• Overcome training barriers:
- Willingness to invest time: Participants indicate how likely they are to spend additional time outside of working hours on a training program they consider important.
- Willingness to bear costs: Participants indicate how likely they are to personally cover the costs of a training program.
- Willingness to seek employer support: Participants indicate how likely they are to request the necessary time and support from their employer, even if such support is not typically provided.

The moderating influence of economic preferences (including self-efficacy, loss aversion, time and risk preferences, procrastination)

Personality and Economic Preferences
To examine potential moderators, the survey includes validated scales to measure individual traits and preferences that may influence how participants respond to the intervention. These include:
• Self-efficacy (Doll et al., 2021)
• Loss aversion (Cabedo-Peris et al., 2024; Duong et al., 2024)
• Time and risk preferences (Falk et al., 2023)
• Procrastination (Börsch-Supan et al., 2023; Tuckman, 1991)

The proposed information experiment will target a sample of 1,200 Dutch-speaking individuals focusing on low- to medium-educated participants who are either employed, seeking employment, or open to job change and development opportunities. Participants will be randomized into three groups (control, gain-framed and loss-framed information treatment), with 400 individuals assigned to each group. Randomization will take place at the individual level to ensure balanced group allocation, and the survey will be administered online to minimize potential spillover effects. The information treatment will be delivered through the survey.

Please note:
- The survey information experiment will run via a Panel Agency.
- Unlike the previous study using the same information intervention (the field experiment) which was conducted within a large employment agency targeting flexible workers, this study will use a broader panel to assess the generalizability of the results to a new population.

Randomization via Survey Software (Qualtrics)

Individual

No

no clusters

1200 individuals (400 per group, 3 groups)

no clusters - 400 per group